I
AE
S
I
nte
rna
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
6
,
No
.
1
,
Ma
r
ch
2
0
1
7
,
p
p
.
33
~
4
8
I
SS
N:
2252
-
8938
,
DOI
: 1
0
.
1
1
5
9
1
/i
j
ai.
v
6
.
i1
.
p
p
33
-
48
33
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
jo
u
r
n
a
l.c
o
m/o
n
lin
e/in
d
ex
.
p
h
p
/I
J
AI
Ti
m
e
-
Ba
sed Ra
g
a
Reco
mm
enda
tio
n and
Infor
m
a
tio
n Ret
riev
a
l
o
f
M
usica
l P
a
t
t
er
ns in India
n
Cla
ss
ica
l Mus
ic U
sing
Neura
l
Netw
o
rk
s
Sa
m
a
rj
it
Ro
y
1
,
Su
dip
t
a
Cha
k
ra
ba
rt
y
2
,
Deba
s
hi
s
De
3
1
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
T
e
c
h
n
o
I
n
d
ia
S
il
li
2
De
p
a
rtme
n
t
o
f
M
CA
,
Tec
h
n
o
In
d
ia S
a
lt
L
a
k
e
3
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter S
c
ien
c
e
a
n
d
E
n
g
in
e
e
rin
g
,
W
e
st Ben
g
a
l
Un
iv
e
rsit
y
o
f
Tec
h
n
o
lo
g
y
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
No
v
6
,
2
0
1
6
R
ev
i
s
ed
J
an
1
0
,
2
0
1
7
A
cc
ep
ted
Feb
1
9
,
2
0
1
7
In
In
d
ian
Clas
sic
a
l
M
u
sic
(IC
M
)
p
e
rsp
e
c
ti
v
e
,
Ra
g
a
is
f
o
r
m
e
d
f
ro
m
th
e
d
if
fe
re
n
t
a
n
d
c
o
rre
c
t
c
o
m
b
in
a
ti
o
n
o
f
n
o
tes
.
If
it
is
o
b
se
rv
e
d
t
h
e
h
isto
ry
o
f
In
d
ian
Clas
sic
a
l
Ra
g
a
in
ICM
,
t
h
e
p
lay
in
g
o
r
se
rv
in
g
e
a
c
h
o
f
th
e
ra
g
a
s
h
a
s
so
m
e
u
n
iq
u
e
se
ss
io
n
s.
T
h
e
p
r
o
c
e
d
u
re
is
t
o
su
g
g
e
st
th
e
c
las
sif
ica
ti
o
n
s
o
f
p
lay
in
g
a
ra
g
a
h
a
s
b
e
e
n
a
tt
e
m
p
t
e
d
to
d
is
p
lay
b
y
e
x
p
lain
in
g
u
n
iq
u
e
m
u
sic
a
l
f
e
a
tu
re
s
a
n
d
p
a
tt
e
rn
m
a
t
c
h
in
g
.
Th
is
c
o
n
tri
b
u
t
io
n
h
a
s
b
e
e
n
re
p
re
se
n
ted
h
o
w
m
u
sic
stru
c
tu
re
s
c
a
n
b
e
a
d
v
a
n
c
e
d
th
ro
u
g
h
a
m
o
re
c
o
n
c
e
p
tu
a
l
d
e
m
o
n
stra
ti
o
n
a
n
d
c
o
n
se
n
t
to
u
n
a
m
b
ig
u
o
u
s
ly
d
e
sc
rib
e
p
ro
c
e
ss
o
f
c
o
m
p
u
tatio
n
a
l
m
o
d
e
li
n
g
o
f
M
u
sic
o
lo
g
y
w
h
ich
sig
n
ify
th
e
c
h
a
ll
e
n
g
e
o
n
c
o
m
p
lete
m
u
sic
a
l
c
o
m
p
o
siti
o
n
f
ro
m
th
e
e
le
m
e
n
tar
y
v
o
c
a
l
o
b
jec
ts
o
f
ICM
u
sa
g
e
u
sin
g
Ne
u
ra
l
Ne
tw
o
rk
s.
In
Ne
u
ra
l
n
e
tw
o
rk
th
e
sa
m
p
les
o
f
v
a
rio
u
s
ra
g
a
s
h
a
v
e
b
e
e
n
ta
k
e
n
a
s
in
p
u
t
a
n
d
c
las
si
fy
th
e
m
a
c
c
o
rd
in
g
to
t
h
e
ti
m
e
s
o
f
th
e
p
e
rf
o
r
m
a
n
c
e
.
Ov
e
r
9
0
%
a
c
c
u
ra
c
y
lev
e
l
h
a
s
a
c
h
iev
e
d
u
sin
g
e
n
ti
re
Co
n
f
u
sio
n
M
a
tri
c
e
s
a
n
d
Err
o
r
Histo
g
ra
m
p
e
rf
o
r
m
a
n
c
e
e
v
a
lu
a
ti
o
n
tec
h
n
iq
u
e
.
K
ey
w
o
r
d
:
Mu
s
ic
Neu
r
al
Net
w
o
r
k
s
No
tes
P
itch
R
ag
a
Co
p
y
rig
h
t
©
2
0
1
7
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Sa
m
ar
j
it R
o
y
,
Dep
ar
t
m
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
i
n
ee
r
in
g
,
T
ec
h
n
o
I
n
d
ia
Sil
li
,
Sil
li,
R
a
n
c
h
i
-
8
3
5
1
0
2
,
J
h
ar
k
h
an
d
,
I
n
d
ia
.
E
m
ail: sa
m
ar
j
it.tec
h
8
9
@
g
m
ail
.
co
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
R
ag
a
in
I
n
d
ian
C
las
s
ical
m
u
s
ic
is
e
x
p
lai
n
ed
as
th
e
e
x
tr
ao
r
d
in
ar
y
a
m
al
g
a
m
atio
n
o
f
th
e
s
et
o
f
n
o
tes
ar
r
an
g
ed
in
a
u
n
iq
u
e
s
eq
u
en
ce
w
h
ic
h
h
as
b
ee
n
cr
ea
ted
th
e
u
n
d
er
g
o
i
n
g
m
u
s
ic
al
e
m
o
tio
n
to
th
e
au
d
ien
ce
s
.
I
f
all
th
e
r
ag
as
an
d
th
er
e
co
r
r
esp
o
n
d
in
g
f
ea
tu
r
e
s
w
il
l
b
e
co
n
te
m
p
lated
in
o
b
j
ec
t
-
o
r
ien
ted
m
a
n
n
er
,
th
en
it
m
u
s
t
b
e
b
elo
n
g
ed
to
s
u
b
-
clas
s
u
n
d
er
a
p
ar
en
t
o
r
s
u
p
er
class
w
h
ic
h
is
k
n
o
w
n
a
s
‘
T
h
h
at
’
i
n
I
n
d
ia
n
C
las
s
ical
m
u
s
ic.
T
h
e
T
h
h
at
is
n
o
th
in
g
b
u
t
t
h
e
n
u
m
er
o
u
s
d
is
tr
ib
u
tio
n
s
o
f
m
u
s
ical
n
o
tes
an
d
t
h
e
n
o
te
-
s
tr
u
ct
u
r
es.
I
n
I
n
d
ia
n
C
lass
ical
Mu
s
ic,
th
er
e
ar
e
te
n
t
h
h
ats
an
d
t
w
el
v
e
n
o
tes.
Am
o
n
g
th
e
t
welv
e
n
o
tes
t
h
er
e
ar
e
s
ev
e
n
p
u
r
e
n
o
tes
an
d
f
i
v
e
s
cr
atch
ed
n
o
te
s
a
v
ailab
le.
E
ac
h
t
h
h
at
ex
ac
tl
y
co
n
s
is
t
s
o
f
s
e
v
e
n
n
o
tes
o
u
t
o
f
e
n
tire
t
w
el
v
e
n
o
te
s
an
d
th
e
n
o
tes
ar
e
u
s
ed
to
b
e
ar
r
an
g
ed
in
ascen
d
in
g
o
r
d
er
s
.
A
ll
t
h
e
r
ag
as
m
u
s
t
b
elo
n
g
to
an
y
o
n
e
o
f
th
e
te
n
th
h
at
s
.
T
h
e
Stru
ct
u
r
es
o
f
t
h
e
n
o
tes
i
n
a
p
ar
ticu
lar
r
ag
a
ca
n
b
e
s
h
o
w
n
d
is
cr
e
p
an
cies
d
u
e
to
th
e
f
r
u
it
f
u
l
au
d
ib
ili
t
y
to
th
e
a
u
d
ien
ce
s
an
d
v
ar
ia
n
t
s
o
f
th
e
s
i
n
g
er
s
o
r
v
o
ca
lis
ts
as
w
el
l
n
e
v
er
th
eles
s
n
u
m
er
o
u
s
n
o
tes
w
i
ll a
l
w
a
y
s
b
e
i
n
tact
to
i
n
to
n
e
b
y
w
h
ich
n
o
te
o
r
s
et
o
f
n
o
tes e
x
ac
t r
ag
a
o
r
it
s
co
r
r
esp
o
n
d
in
g
T
h
h
ats
h
a
v
e
b
ee
n
ca
teg
o
r
ized
.
T
h
e
p
r
ev
i
o
u
s
h
is
to
r
y
o
f
I
n
d
ian
m
u
s
ic
s
aid
th
e
ex
p
lici
t
ex
p
la
n
atio
n
s
ab
o
u
t
th
e
r
ag
a
ch
a
r
ac
ter
is
tic
s
an
d
th
e
ac
t
u
al
t
i
m
i
n
g
o
f
p
la
y
i
n
g
th
e
r
ag
a
s
.
T
h
e
Neu
r
al
n
et
w
o
r
k
s
ar
e
s
el
f
-
p
o
s
s
es
s
ed
o
f
tr
o
u
b
le
-
f
r
ee
ele
m
en
ts
w
o
r
k
in
g
i
n
a
n
alo
g
o
u
s
s
y
s
te
m
s
o
r
ess
e
n
tials
.
T
h
ese
ele
m
e
n
t
s
ar
e
h
ab
itu
all
y
m
o
ti
v
ated
b
y
b
io
lo
g
ical
n
er
v
o
u
s
s
y
s
te
m
s
an
d
as
d
esig
n
s
an
d
op
er
atio
n
s
p
er
s
p
ec
tiv
es,
th
e
y
ar
e
s
i
m
ilar
to
th
e
o
p
er
atio
n
s
o
f
h
u
m
a
n
b
ein
g
s
.
As
in
n
atu
r
e,
th
e
n
et
w
o
r
k
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
6
,
No
.
1
,
Ma
r
ch
2
0
1
7
:
33
–
4
8
34
f
u
n
ctio
n
is
d
e
f
in
ed
p
r
in
cip
all
y
b
y
t
h
e
ass
o
ciat
io
n
s
a
m
o
n
g
ele
m
e
n
ts
.
A
n
eu
r
al
n
et
w
o
r
k
ca
n
b
e
tr
ain
ed
to
p
er
f
o
r
m
a
p
ar
ticu
lar
f
u
n
ctio
n
b
y
a
m
e
n
d
i
n
g
th
e
id
ea
ls
o
f
t
h
e
co
n
n
ec
t
io
n
s
n
a
m
el
y
w
eig
h
t
s
b
et
w
ee
n
ele
m
e
n
ts
.
Fre
q
u
en
tl
y
n
eu
r
al
n
et
w
o
r
k
s
ar
e
f
a
m
iliar
o
r
tr
ain
ed
,
as
i
f
a
p
ar
ticu
lar
in
p
u
t
esco
r
ts
to
a
p
r
ec
is
e
tar
g
et
o
u
tp
u
t.
Su
c
h
a
s
i
tu
atio
n
h
as
b
ee
n
s
h
o
w
n
i
n
to
t
h
is
co
n
tr
ib
u
tio
n
b
elo
w
.
I
n
t
h
i
s
co
n
tr
ib
u
tio
n
t
h
e
n
e
t
w
o
r
k
is
ad
j
u
s
ted
,
d
ep
en
d
in
g
u
p
o
n
a
co
m
p
ar
is
o
n
o
f
t
h
e
p
r
o
d
u
ctio
n
a
n
d
t
h
e
tar
g
et,
u
n
t
il
t
h
e
n
et
w
o
r
k
o
u
t
p
u
t
is
eq
u
i
v
ale
n
t
to
th
e
tar
g
et.
Her
e
in
th
is
co
n
tr
ib
u
tio
n
,
t
h
e
ti
m
e
-
b
ased
r
ag
a
r
ec
o
m
m
e
n
d
at
io
n
an
d
m
o
d
elin
g
o
n
th
e
ac
t
u
al
p
lay
i
n
g
ti
m
e
o
f
r
a
g
a
ar
e
an
al
y
ze
d
.
T
h
e
r
ag
as
ar
e
class
i
f
ied
w
it
h
th
e
h
elp
o
f
n
e
u
r
al
n
et
w
o
r
k
s
w
h
ic
h
d
iv
id
es
t
h
e
s
et
o
f
r
ag
a
s
a
m
p
les
i
n
to
t
w
o
cla
s
s
e
s
o
f
t
i
m
e
s
o
r
s
es
s
io
n
s
o
f
p
er
f
o
r
m
a
n
ce
s
n
a
m
el
y
,
‘
P
u
r
v
an
g
o
’
(
f
r
o
m
1
2
p
m
to
1
2
a
m
)
an
d
‘
U
ttar
an
g
o
’
(
f
r
o
m
1
2
a
m
to
1
2
p
m
)
.
T
h
e
i
n
f
o
r
m
atio
n
p
r
o
ce
s
s
i
n
g
n
e
u
r
o
n
s
o
f
th
e
n
e
u
r
al
n
et
w
o
r
k
ar
e
tr
ain
ed
b
y
th
e
d
ata
s
et
a
n
d
th
en
tes
ted
b
y
a
n
u
m
b
er
o
f
s
a
m
p
les.
T
h
e
tr
ai
n
i
n
g
p
r
o
ce
d
u
r
e
h
as
b
ee
n
co
n
ti
n
u
ed
till
th
e
o
p
ti
m
u
m
ac
c
u
r
ac
y
o
cc
u
r
r
ed
o
f
th
e
ex
ec
u
tio
n
.
Du
e
to
s
e
v
er
al
d
is
p
ar
ities
i
n
n
o
tes
s
tr
u
ct
u
r
es
u
s
u
all
y
d
en
o
te
d
in
I
n
d
ian
C
las
s
ical
M
u
s
ic
f
o
r
d
ef
in
i
n
g
th
e
t
h
h
a
ts
t
h
e
u
n
iq
u
e
c
h
ar
ac
ter
is
tics
o
f
r
ag
a
s
al
s
o
m
u
s
t b
e
f
l
u
ctu
ated
.
Her
e
i
n
t
h
is
co
n
tr
ib
u
ti
o
n
,
th
e
t
i
m
e
-
b
ase
d
r
ag
a
r
ec
o
m
m
e
n
d
atio
n
a
n
d
m
o
d
elin
g
o
n
t
h
e
ac
t
u
al
p
la
y
i
n
g
ti
m
e
o
f
r
ag
a
ar
e
an
a
l
y
ze
d
.
2.
RE
L
AT
E
D
WO
RK
S
C
o
m
p
u
ter
-
b
ased
ex
p
lo
r
atio
n
o
n
f
u
n
d
a
m
en
ta
l
m
u
s
ical
f
ea
t
u
r
es,
th
eir
co
r
r
esp
o
n
d
in
g
r
h
y
t
h
m
ic
c
y
cle
s
an
d
attr
ib
u
te
s
o
f
C
las
s
ical
m
u
s
ic
h
as
b
ee
n
r
elati
v
el
y
s
p
o
r
ad
ic,
an
d
u
p
to
th
i
s
p
o
in
t
o
f
v
ie
w
r
e
m
ar
k
ab
le
atte
m
p
ts
h
a
v
e
b
ee
n
m
ad
e
to
ca
teg
o
r
ize
t
h
e
v
o
ca
l
a
n
d
r
h
y
t
h
m
ic
f
o
r
m
atio
n
s
b
y
d
es
ig
n
.
On
t
h
e
co
n
tr
ar
y
,
th
e
r
ec
o
g
n
itio
n
o
f
t
h
e
p
atter
n
s
i
n
Hin
d
u
s
ta
n
i
C
lass
ical
R
a
g
a
s
h
a
s
b
ee
n
a
n
i
n
ti
m
ate
ar
ea
w
h
er
e
s
ev
er
al
c
h
alle
n
g
es
h
av
e
b
ee
n
p
r
ep
ar
ed
to
p
e
r
ce
iv
e
a
q
u
an
tit
y
o
f
d
is
ti
n
ctio
n
a
m
o
n
g
th
e
r
ag
a
b
lu
ep
r
in
ts
an
d
an
alo
g
o
u
s
r
h
y
t
h
m
ic
f
ea
t
u
r
es.
I
n
a
p
ap
er
,
th
e
r
esear
ch
er
s
d
e
m
o
n
s
tr
ated
a
s
u
r
v
e
y
o
n
co
m
p
u
tatio
n
a
ll
y
s
u
p
p
o
r
ted
m
u
s
ica
l
co
m
p
o
s
i
tio
n
b
y
f
o
c
u
s
i
n
g
o
n
P
etr
i
Nets
an
d
g
av
e
s
ev
er
al
m
aj
o
r
p
ath
s
o
f
th
eir
ap
p
licatio
n
s
.
On
th
is
b
asis
i
n
ter
m
s
o
f
P
etr
i
n
ets
th
is
ef
f
o
r
ts
p
r
o
v
id
ed
in
n
o
v
ati
v
e
s
tep
s
f
o
r
m
ak
i
n
g
th
e
s
e
lab
o
r
s
o
p
er
atio
n
al
in
co
n
cr
ete
co
m
p
u
ter
ized
e
n
v
ir
o
n
m
e
n
t
s
,
n
o
t
j
u
s
t
at
th
e
le
v
el
o
f
f
o
r
m
al
ab
s
tr
ac
tio
n
[
1
]
.
T
h
e
au
t
h
o
r
s
h
av
e
elab
o
r
ated
th
e
alter
n
ati
v
e
ap
p
r
o
ac
h
es
f
o
r
ex
p
lo
r
in
g
a
n
d
r
elati
n
g
to
t
h
e
d
esi
g
n
an
d
f
u
n
c
tio
n
o
f
t
h
e
s
o
f
t
w
a
r
e
p
r
o
j
ec
t
"
No
d
al
"
.
T
h
is
p
ar
ticu
lar
w
o
r
k
ai
m
s
to
cr
ea
te
a
g
r
ap
h
ical
en
v
ir
o
n
m
en
t
t
h
at
en
ab
le
s
th
e
u
s
er
to
co
n
f
i
g
u
r
e
a
s
p
atial,
d
ir
ec
ted
g
r
ap
h
th
at
g
en
er
ate
s
m
u
s
ic
in
r
ea
l
-
ti
m
e
s
y
s
te
m
s
.
T
h
ey
r
ef
er
r
ed
to
s
u
ch
g
r
ap
h
s
as
co
m
p
o
s
itio
n
o
r
n
o
d
al
n
et
w
o
r
k
s
.
T
h
e
d
i
s
cu
s
s
io
n
in
t
h
is
w
o
r
k
h
a
s
b
ee
n
r
elate
d
to
th
e
f
u
n
d
a
m
en
tal
d
esig
n
co
n
s
tr
ain
ts
t
h
at
h
a
v
e
b
ee
n
i
m
p
o
s
ed
w
it
h
i
n
t
h
ese
co
n
s
tr
ai
n
ts
t
h
at
g
i
v
e
r
is
e
to
d
i
f
f
er
en
t
m
u
s
ical
b
eh
a
v
io
r
s
[
2
]
.
I
n
a
n
o
th
er
p
iece
o
f
w
o
r
k
s
t
h
e
a
u
t
h
o
r
h
a
s
d
escr
ib
e
d
an
d
r
esear
ch
ed
o
n
t
h
e
co
m
p
u
ta
tio
n
a
l
m
u
s
ic
th
eo
r
y
w
h
ich
h
as b
ee
n
e
s
tab
lis
h
ed
b
y
J
o
h
n
C
lo
u
g
h
.
A
cc
o
r
d
in
g
to
th
e
t
h
o
u
g
h
t o
f
C
o
m
p
u
tatio
n
al
Mu
s
ic,
t
h
is
h
as
b
ee
n
r
e
s
o
lu
te
th
at
t
h
e
r
esear
c
h
o
f
J
o
h
n
C
lo
u
g
h
w
as
n
o
t
o
n
l
y
r
estricte
d
to
th
e
m
at
h
e
m
atic
al
th
eo
r
y
b
u
t
al
s
o
s
u
r
r
o
u
n
d
e
d
in
s
cien
ce
s
an
d
h
u
m
a
n
itie
s
to
o
.
T
h
e
au
t
h
o
r
h
as
ill
u
s
tr
ated
al
s
o
t
h
e
s
p
ec
i
f
ic
an
d
g
e
n
er
ic
i
n
ter
m
is
s
io
n
w
h
i
ch
is
k
n
o
w
n
as
th
e
M
y
h
ill
’
s
p
r
o
p
er
ty
[
3
]
.
T
h
e
r
esear
ch
er
s
h
av
e
a
f
f
o
r
d
ed
th
e
p
er
f
o
r
m
an
ce
o
f
m
u
s
ic
ca
n
i
m
p
r
o
v
e
th
e
b
eh
a
v
io
r
al
p
er
f
o
r
m
a
n
ce
s
li
k
e
t
h
e
h
u
m
a
n
in
tel
lig
e
n
ce
s
y
s
te
m
s
i
n
b
o
th
s
tr
u
c
tu
r
al
a
n
d
f
u
n
ct
io
n
al
al
ti
tu
d
es.
A
p
ar
ticu
lar
m
u
s
ic
ca
n
lead
to
th
i
n
k
an
lis
ten
er
to
i
m
p
r
o
v
e
s
o
m
e
b
eh
a
v
i
o
r
d
o
m
ain
s
o
f
m
u
s
ic
l
ik
e
t
h
e
v
o
ca
l
p
er
f
o
r
m
a
n
ce
n
a
m
e
l
y
t
h
e
lan
g
u
ag
e
t
h
at
m
ig
h
t
b
e
m
o
r
e
f
r
u
it
f
u
l.
T
h
e
au
th
o
r
s
h
av
e
also
ex
p
lo
r
ed
th
e
f
u
t
u
r
e
tr
en
d
ab
o
u
t
th
e
in
ter
r
elatio
n
s
h
ip
s
a
m
o
n
g
m
u
s
ic
an
d
t
h
e
la
n
g
u
a
g
es
[
4
]
.
T
h
e
au
t
h
o
r
s
i
n
t
h
eir
p
ap
er
[
5
]
h
av
e
m
en
tio
n
ed
t
h
e
m
u
s
ic
in
t
h
e
p
ed
ag
o
g
y
o
f
m
at
h
e
m
a
tics
w
h
ich
h
a
s
b
ee
n
d
ev
e
lo
p
e
d
b
y
Ma
t
h
e
m
atic
al
Mu
s
ic
T
h
eo
r
y
.
T
h
e
f
u
n
d
a
m
e
n
tal
in
ten
s
io
n
o
f
th
e
a
u
t
h
o
r
s
is
e
lab
o
r
ated
to
p
o
p
u
lar
ize
th
e
m
u
s
ical
p
r
o
j
ec
t b
r
o
ad
ly
Ma
th
e
m
atic
s
an
d
Mu
s
ic
f
o
r
m
t
h
e
r
e
g
io
n
al
c
u
lt
u
r
e
to
in
ter
n
atio
n
al
le
v
el.
T
h
e
y
a
ls
o
v
ie
w
to
cr
ea
te
t
h
e
d
i
d
ac
tic
m
ater
ial
s
b
y
w
h
ic
h
all
ca
n
g
a
th
er
t
h
e
n
o
v
el
id
ea
s
a
n
d
s
k
etch
e
s
ab
o
u
t
Ma
th
e
m
atica
l
M
u
s
ic
t
h
eo
r
y
.
A
d
d
itio
n
a
ll
y
,
t
h
e
r
esear
ch
er
s
e
x
er
cised
t
h
e
m
et
h
o
d
o
f
ti
m
e
s
er
ies
a
n
al
y
s
i
s
to
co
m
p
ar
e
th
e
e
x
p
er
tis
e
g
r
o
u
p
s
an
d
i
n
d
i
v
id
u
al
s
i
n
d
y
n
a
m
ic
u
n
r
e
m
i
tti
n
g
d
is
ce
r
n
m
en
t
o
f
ar
o
u
s
a
l
i
n
m
u
s
ic.
Fo
r
test
i
n
g
t
h
e
v
alid
it
y
t
h
e
y
u
s
e
d
th
e
g
e
n
er
al
li
n
ea
r
au
to
r
eg
r
es
s
iv
e
m
o
v
in
g
a
v
er
ag
e
[
6
]
.
T
h
e
au
th
o
r
s
h
a
v
e
ev
id
en
tl
y
r
ep
r
esen
ted
t
h
e
e
x
p
licatio
n
an
d
d
is
c
u
s
s
io
n
o
f
m
et
h
o
d
o
lo
g
ies
i
n
co
m
p
u
tatio
n
al,
m
at
h
e
m
atica
l
a
n
d
s
tat
is
tica
l
m
u
s
ic
ap
p
r
o
ac
h
[
7
]
.
T
h
e
au
t
h
o
r
s
h
av
e
also
h
er
e
d
is
cu
s
s
ed
th
e
in
ter
-
a
s
s
o
ciatio
n
o
f
m
u
s
ic
w
i
th
s
cie
n
ce
,
co
g
n
it
iv
e
s
c
ien
ce
a
n
d
h
u
m
a
n
itie
s
.
I
n
a
d
is
s
er
tatio
n
th
e
r
ag
a
h
as
b
ee
n
ap
p
r
o
ac
h
ed
an
d
id
en
tif
ied
to
r
ec
o
g
n
ize
f
r
o
m
a
r
eg
io
n
al
m
u
s
ic
p
er
f
o
r
m
a
n
ce
l
ik
e
C
ar
n
a
tic
m
u
s
ic
s
ig
n
al.
T
h
e
m
et
h
o
d
h
a
s
b
ee
n
d
ep
lo
y
ed
to
s
ep
ar
ate
th
e
v
o
ca
l
d
ata
an
d
in
s
tr
u
m
e
n
tal
p
er
f
o
r
m
a
n
ce
s
f
r
o
m
a
p
o
ly
p
h
o
n
ic
m
u
s
ic
s
ig
n
al
th
r
o
u
g
h
s
i
g
n
a
l
s
ep
ar
atio
n
alg
o
r
ith
m
.
A
cc
o
r
d
in
g
to
o
n
l
y
th
e
v
o
c
al
d
ata
r
etr
iev
al,
th
e
n
o
tes
h
av
e
b
ee
n
r
ec
o
g
n
ized
b
y
t
h
eir
in
i
tial
f
r
eq
u
en
c
ies
[
8
]
.
I
n
a
co
n
tr
ib
u
t
io
n
t
h
e
m
u
s
ical
e
m
o
tio
n
r
ec
o
g
n
itio
n
s
y
s
te
m
h
a
s
b
ee
n
p
r
esen
t
d
ev
elo
p
ed
b
y
t
h
e
o
n
e
-
cla
s
s
-
in
-
o
n
e
p
h
e
n
o
m
e
n
a
o
f
n
eu
r
al
n
et
w
o
r
k
s
[
9
]
.
Her
e
th
e
s
y
s
te
m
i
s
en
tire
l
y
s
p
ea
k
er
an
d
co
n
tex
t
-
i
n
d
ep
en
d
en
t.
A
d
d
itio
n
all
y
,
s
tr
in
g
r
e
w
r
it
in
g
g
r
a
m
m
ar
s
b
ased
o
n
L
-
S
y
s
te
m
s
h
a
v
e
b
ee
n
ad
o
p
te
d
in
to
a
h
ar
m
o
n
izatio
n
f
o
r
m
u
s
ic
m
aster
p
iece
s
.
I
llu
s
tr
ati
o
n
o
f
p
itch
p
er
io
d
s
,
in
ter
v
a
l
a
n
d
ti
m
b
r
e
ar
e
p
r
ea
r
r
an
g
ed
a
s
g
r
a
m
m
ar
cr
y
p
to
g
r
a
m
,
u
p
o
n
w
h
ic
h
s
eq
u
en
ce
o
f
r
e
-
w
r
iti
n
g
r
u
le
s
ar
e
ap
p
lied
[
1
0
]
.
T
h
e
s
y
s
te
m
p
r
o
v
is
io
n
s
th
e
d
ata
alo
n
g
w
it
h
i
n
ce
p
tio
n
ti
m
e
s
in
f
i
g
u
r
ativ
e
s
co
r
e
w
h
ic
h
h
a
s
u
s
ed
f
o
r
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
Time
-
B
a
s
ed
R
a
g
a
R
ec
o
mme
n
d
a
tio
n
a
n
d
I
n
fo
r
ma
tio
n
R
etri
ev
a
l o
f..
.
(
S
a
ma
r
jit R
o
y
)
35
alter
atio
n
a
n
d
h
ab
itu
a
l
d
ictatio
n
i
n
s
o
lo
p
er
f
o
r
m
a
n
ce
in
T
ab
la
[
1
1
]
.
A
b
o
u
t
th
e
attr
ib
u
te
as
s
o
r
tm
e
n
t
an
d
m
in
i
n
g
in
s
tr
u
m
en
t
f
r
o
m
t
h
e
u
n
u
s
u
al
v
ar
iet
y
o
f
m
u
s
ical
in
s
tr
u
m
e
n
t
s
th
a
t
w
o
r
n
a
s
t
h
e
s
t
u
f
f
in
m
u
s
ic
is
s
y
m
b
o
lized
i
n
class
p
air
-
w
is
e
co
m
p
o
r
t
m
e
n
t
[
1
2
]
.
A
d
d
itio
n
all
y
in
s
e
v
er
al
e
f
f
o
r
t
s
b
y
th
e
au
t
h
o
r
s
h
av
e
b
ee
n
d
escr
ib
ed
th
e
m
u
s
ical
p
atter
n
r
ec
o
g
n
i
tio
n
a
n
d
r
h
y
t
h
m
ic
f
ea
t
u
r
es
e
x
p
o
s
u
r
es
b
y
o
b
j
ec
t
-
o
r
ien
ted
m
a
n
n
er
[
1
3
-
1
4
]
.
A
lo
t
o
f
d
is
cu
s
s
io
n
s
ar
e
a
v
ailab
le
ab
o
u
t
th
e
f
ea
t
u
r
es
cla
s
s
i
f
ica
tio
n
u
s
i
n
g
UM
L
-
o
r
ie
n
ted
clas
s
d
iag
r
a
m
s
a
n
d
t
h
e
w
a
y
o
f
i
m
p
le
m
en
ta
tio
n
s
o
f
m
u
s
ical
r
h
y
t
h
m
ic
c
y
c
les
f
o
r
th
e
p
er
cu
s
s
io
n
-
b
ased
in
s
tr
u
m
e
n
ts
[
1
5
-
1
6
]
.
T
h
e
au
to
m
ated
r
h
y
t
h
m
g
e
n
er
atio
n
i
s
o
n
e
o
f
t
h
e
h
ar
d
est
p
r
o
b
le
m
ar
ea
s
.
T
h
is
is
b
ec
au
s
e
o
f
co
n
s
tr
ain
t
s
a
n
d
s
atis
f
y
i
n
g
t
h
o
s
e
co
n
s
tr
ain
ts
to
g
et
th
e
f
ea
s
ib
le
an
d
o
p
ti
m
iz
ed
r
h
y
t
h
m
.
I
n
an
e
f
f
o
r
t
b
y
t
h
e
a
u
t
h
o
r
s
h
av
e
b
ee
n
m
o
d
eled
(
Qu
alit
y
Me
a
s
u
r
e
m
o
d
el)
a
n
o
p
tim
ized
al
g
o
r
ith
m
o
n
r
h
y
t
h
m
ic
c
y
cle
s
u
s
ed
i
n
I
n
d
ia
n
C
l
ass
ical
Mu
s
ic
u
s
in
g
th
e
R
o
u
let
te
-
W
h
ee
l
Se
lectio
n
o
f
Ge
n
etic
Alg
o
r
it
h
m
C
o
n
ce
p
t
[
1
8
]
.
Sev
er
al
p
ap
er
s
h
a
v
e
b
ee
n
co
n
s
id
e
r
ed
th
o
s
e
ar
e
b
ased
o
n
t
h
e
cr
ea
ti
v
it
y
o
f
m
u
s
ic
u
s
in
g
Gen
etic
A
l
g
o
r
ith
m
tec
h
n
iq
u
e.
Fu
r
t
h
er
s
o
m
e
o
t
h
er
p
ap
er
s
d
ea
l
w
it
h
th
e
i
m
p
le
m
en
ta
tio
n
o
f
M
u
s
ic
al
p
atter
n
r
ec
o
g
n
itio
n
b
y
m
a
th
e
m
atica
l
e
x
p
r
ess
io
n
s
.
T
h
e
y
s
u
g
g
ested
a
n
e
w
m
et
h
o
d
f
o
r
ar
r
an
g
in
g
d
i
f
f
er
e
n
t
m
elo
d
io
u
s
a
u
d
io
s
tr
ea
m
i
n
to
s
o
m
e
s
p
ec
if
ic
t
y
p
e
o
f
clas
s
es
b
ased
o
n
o
b
j
ec
t
o
r
ien
ted
p
ar
ad
ig
m
.
I
n
a
p
ap
er
,
th
e
au
th
o
r
s
h
a
v
e
ex
p
lai
n
ed
elab
o
r
ately
t
h
e
n
o
tes
an
d
th
e
n
o
tes
s
tr
u
ctu
r
es
a
s
w
ell
as
t
h
e
te
n
th
h
at
s
o
r
r
ag
a
o
r
ig
in
in
I
n
d
ia
n
cla
s
s
ica
l
m
u
s
i
c
w
h
ic
h
is
i
m
p
o
s
ed
to
b
e
co
n
s
tr
u
cted
t
h
e
s
o
n
g
s
in
I
n
d
ian
o
r
Hin
d
u
s
tan
i
m
u
s
ica
l
p
atter
n
s
.
T
h
e
alg
o
r
ith
m
h
a
s
b
ee
n
d
ef
i
n
ed
h
er
e
w
h
ic
h
ca
n
d
et
er
m
in
e
t
h
e
t
h
h
at
o
r
r
ag
a
o
r
ig
in
i
n
w
h
ich
a
s
o
n
g
in
I
n
d
ia
n
m
u
s
ic
b
elo
n
g
s
[
1
7
]
.
So
m
e
r
esear
c
h
er
s
in
tr
o
d
u
ce
a
s
y
s
te
m
f
o
r
r
ec
o
g
n
itio
n
o
f
t
h
e
p
atter
n
s
o
f
m
u
s
ic
b
y
u
s
in
g
k
e
y
s
eq
u
en
ce
s
w
it
h
R
ec
u
r
s
i
v
e
Me
d
ia
n
Fil
ter
[
1
9
]
.
Sev
er
al
co
n
tr
ib
u
tio
n
s
h
av
e
b
ee
n
d
ep
icted
u
p
o
n
au
d
io
o
r
s
p
ee
ch
r
ec
o
g
n
itio
n
s
y
s
te
m
s
w
h
ich
ar
e
co
n
ten
t
b
ased
,
f
ea
t
u
r
ed
-
b
ased
p
ar
a
m
eter
izati
o
n
f
o
r
m
u
s
ic
i
n
f
o
r
m
atio
n
r
etr
iev
al
u
s
i
n
g
Ne
u
r
al
Net
w
o
r
k
s
a
p
p
r
o
ac
h
as
w
ell
[
2
2
-
2
3
,
2
5
-
2
6
]
I
n
an
e
f
f
o
r
t
t
h
e
au
d
io
ch
ar
ac
ter
is
t
ics
al
s
o
m
o
d
eled
th
o
s
e
ar
e
e
m
a
n
ated
f
r
o
m
th
e
h
ea
r
i
n
g
aid
s
u
s
i
n
g
Neu
r
al
Net
w
o
r
k
s
w
it
h
u
s
er
in
p
u
t
[
2
4
]
.
T
h
e
w
o
r
k
s
o
f
th
e
r
esear
c
h
er
s
h
a
v
e
elu
c
id
ated
also
th
e
en
tir
e
m
u
s
ic
co
m
p
o
s
i
tio
n
a
n
d
class
i
f
icatio
n
b
y
n
eu
r
al
ap
p
r
o
ac
h
es
[
2
9
-
3
0
]
as
w
ell
as
t
h
eir
s
e
m
an
tic,
ep
is
o
d
ic
[
2
8
]
an
d
m
u
s
ic
-
e
m
o
tio
n
b
eh
a
v
io
r
[
3
1
-
3
2
]
.
T
h
e
s
u
p
p
o
r
tin
g
th
e
m
es
u
p
o
n
m
u
s
ic
i.e
.
,
in
s
tr
u
m
en
tal
p
er
f
o
r
m
an
ce
s
w
h
ic
h
h
as
al
s
o
b
ee
n
c
lass
if
ied
p
r
o
d
u
ctiv
el
y
[
3
3
]
u
s
i
n
g
t
h
eir
p
er
f
o
r
m
in
g
f
ea
t
u
r
es
[
3
4
,
3
6
]
.
T
h
e
class
ical
m
u
s
ic
o
r
ien
tatio
n
s
h
a
v
e
b
ee
n
d
o
n
e
in
s
ev
er
al
co
n
tr
ib
u
tio
n
s
u
s
in
g
s
o
m
e
m
o
d
eli
n
g
co
n
ce
p
t
li
k
e
P
etr
i
n
et
[
3
8
-
3
9
]
.
T
h
e
co
n
tex
t
-
a
w
ar
e
p
er
v
asiv
e
d
iar
y
f
o
r
m
u
s
ic
co
m
p
u
t
in
g
h
as
b
ee
n
d
is
c
u
s
s
ed
b
y
L
i
n
ea
r
R
an
k
Se
lectio
n
m
ec
h
a
n
i
s
m
o
f
Gen
etic
A
l
g
o
r
ith
m
[
4
0
]
.
Nu
m
er
o
u
s
ap
p
r
o
ac
h
es
ar
e
b
ased
o
n
m
u
s
ica
l
f
ea
t
u
r
es
ex
tr
ac
tio
n
f
r
o
m
th
e
I
n
d
ian
C
las
s
ical
R
a
g
a
o
r
in
s
tr
u
m
en
ta
l
r
h
y
t
h
m
s
[
4
1
-
4
3
]
.
T
h
e
au
t
h
o
r
s
h
a
v
e
e
x
p
licated
th
e
f
o
r
m
a
tio
n
s
o
f
m
u
s
ical
p
iece
s
b
y
f
u
zz
y
lo
g
ic
[
4
5
]
an
d
co
n
s
tr
ain
t
s
b
y
in
teg
r
atio
n
o
f
DSP
alg
o
r
ith
m
s
[
4
6
]
.
T
h
e
co
n
tr
ib
u
tio
n
s
o
f
th
e
r
esear
c
h
er
s
h
av
e
al
s
o
b
ee
n
h
i
g
h
lig
h
ted
to
t
h
e
n
o
v
el
ap
p
r
o
ac
h
es
f
o
r
m
u
s
ic
r
ec
o
m
m
en
d
atio
n
[
4
7
-
4
8
]
.
Qu
ite
a
f
e
w
to
ils
h
a
v
e
clar
if
ied
th
e
ac
tu
a
l
m
u
s
ical
f
ea
tu
r
es
o
f
I
n
d
ian
C
lass
ical
v
o
ca
l
an
d
r
h
y
t
h
m
ic
d
is
tin
ct
iv
e
n
es
s
an
d
t
h
eir
co
r
r
esp
o
n
d
in
g
p
itch
v
al
u
e
a
n
al
y
s
is
[
4
9
-
5
6
]
.
T
h
is
h
as
b
ee
n
s
u
p
p
o
r
tiv
el
y
e
x
p
lain
ed
i
n
th
is
f
r
a
m
e
w
o
r
k
to
t
h
e
a
f
o
r
esai
d
p
r
o
ce
d
u
r
al
m
ec
h
a
n
i
s
m
to
e
x
p
lo
r
e
a
p
io
n
ee
r
in
g
m
o
d
e
o
f
d
e
p
ictio
n
o
f
m
u
s
ical
f
ea
t
u
r
es a
n
d
p
atter
n
m
atc
h
in
g
tr
ial
s
b
ased
o
n
Neu
r
al
Net
w
o
r
k
s
co
n
ce
p
t
s
.
3.
RE
S
E
ARCH
M
E
T
H
O
D
I
n
th
e
p
er
s
p
ec
ti
v
e
f
o
r
cr
ea
tin
g
I
n
d
ia
n
Mu
s
ic,
t
h
er
e
ar
e
b
asicall
y
t
w
o
r
u
d
i
m
e
n
t
s
.
I
t
m
u
s
t
f
o
llo
w
a
n
in
ter
n
a
l
s
k
ele
to
n
li
k
e
I
n
d
ian
C
lass
ica
l
R
a
g
a
an
d
a
s
p
ec
if
ic
r
h
y
t
h
m
.
T
h
e
en
tire
v
o
ca
l
in
g
r
ed
ien
ts
u
s
e
t
o
d
ev
elo
p
b
y
f
o
llo
w
i
n
g
t
h
e
f
u
n
d
am
e
n
tal
r
ag
a
-
f
ea
tu
r
es
o
f
m
u
s
ic.
A
ll
t
h
e
r
a
g
as
ar
e
t
h
e
v
ar
i
atio
n
s
o
f
ele
m
en
tar
y
m
u
s
ical
n
o
te
s
r
ec
o
g
n
izi
n
g
u
n
iq
u
el
y
t
h
e
t
h
h
a
ts
.
A
cc
o
r
d
in
g
to
th
e
s
en
s
e
o
f
m
u
s
ical
t
h
eo
r
y
an
d
th
e
u
n
iq
u
e
ch
ar
ac
ter
is
tic
s
o
f
n
o
te
-
s
tr
u
ct
u
r
es th
e
t
h
h
a
ts
an
d
r
ag
a
s
ar
e
to
b
e
d
ev
elo
p
ed
.
I
n
T
ab
le
1
,
th
e
ac
tu
al
ti
m
e
o
f
s
in
g
i
n
g
,
u
s
ed
u
n
iq
u
e
n
o
te
s
o
n
l
y
o
n
th
e
p
ar
ticu
lar
ti
m
e
d
u
r
ati
o
n
,
th
h
at
s
u
s
ed
o
n
t
h
at
ti
m
e
an
d
s
e
v
er
al
co
r
r
esp
o
n
d
in
g
r
ag
as
ar
e
m
en
t
i
o
n
ed
.
On
th
e
d
u
r
atio
n
o
f
p
la
y
i
n
g
t
h
e
r
ag
a
s
e
ts
o
f
co
m
m
o
n
m
u
s
ica
l
n
o
tes
also
ar
e
in
clu
d
ed
b
u
t
th
e
u
n
iq
u
e
n
o
tes
s
h
o
u
ld
b
e
p
r
esen
t
ac
co
r
d
in
g
to
th
e
tab
le
m
en
tio
n
ed
.
R
ag
a
T
i
m
e
-
C
y
cle
i
s
t
h
e
b
r
o
ad
s
e
n
s
e
o
f
r
a
g
a
-
f
ea
t
u
r
es
an
a
l
y
s
i
s
ac
co
r
d
in
g
to
t
h
e
r
ev
o
l
u
tio
n
o
f
ti
m
e
o
r
s
ess
io
n
.
I
t
i
s
alie
n
ated
w
it
h
t
h
e
ti
m
e.
All
t
h
e
t
w
e
n
t
y
-
f
o
u
r
h
o
u
r
s
o
f
a
d
a
y
is
d
i
v
id
ed
in
to
t
welv
e
f
r
ag
m
e
n
ts
w
it
h
th
e
ev
er
y
t
w
o
h
o
u
r
s
o
f
i
n
ter
v
a
ls
.
Fo
r
‘
u
t
to
r
an
g
o
’
s
e
s
s
io
n
t
h
er
e
ar
e
eig
h
t d
iv
is
io
n
s
an
d
d
u
r
at
io
n
o
f
4
a
m
to
4
p
m
as
Da
w
n
,
E
ar
l
y
Mo
r
n
i
n
g
,
Mo
r
n
in
g
,
L
ate
Mo
r
n
i
n
g
,
Af
ter
n
o
o
n
an
d
L
ate
Af
ter
n
o
o
n
.
I
n
th
i
s
s
ess
io
n
,
all
th
e
ten
u
n
iq
u
e
C
la
s
s
ical
T
h
h
at
s
ar
e
v
ar
ied
w
it
h
t
h
e
f
ea
t
u
r
ed
-
r
ag
a
s
.
Un
iq
u
e
n
o
tes
ar
e
ca
teg
o
r
ized
in
th
e
r
ag
a
-
ti
m
e
tab
le
T
ab
le
1
.
Fo
r
‘
P
u
r
b
an
g
o
’
s
e
s
s
io
n
t
h
er
e
ar
e
ei
g
h
t
d
i
v
is
io
n
s
an
d
d
u
r
atio
n
o
f
4
p
m
to
4
a
m
as
Du
s
k
,
E
v
en
i
n
g
,
L
ate
E
v
e
n
in
g
,
N
ig
h
t
,
Mid
n
i
g
h
t
an
d
P
r
e
-
Da
w
n
.
I
n
th
is
s
e
s
s
io
n
,
n
in
e
o
f
a
ll
t
h
e
te
n
u
n
iq
u
e
C
las
s
ica
l
T
h
h
ats
ar
e
v
ar
ied
w
ith
t
h
e
f
ea
tu
r
ed
-
r
ag
a
s
.
I
n
b
o
th
o
f
th
ese
s
ess
io
n
s
,
v
ar
iatio
n
s
o
f
all
th
e
class
ical
n
o
te
s
r
e
m
ain
al
m
o
s
t
i
n
tact
a
s
w
ell
a
s
al
m
o
s
t
n
i
n
e
th
h
at
s
ar
e
s
a
m
e.
B
o
th
o
f
th
ese
t
w
o
s
e
s
s
io
n
s
co
n
tai
n
t
w
el
v
e
h
o
u
r
s
an
d
s
i
x
f
r
a
g
m
e
n
t
s
.
Ne
v
er
th
e
l
ess
,
t
h
e
r
a
g
a
f
ea
t
u
r
es
a
n
d
u
s
ed
-
r
ag
a
ar
e
d
i
f
f
er
en
t
f
o
r
b
o
th
‘
Utto
r
an
g
o
’
a
n
d
‘
P
u
r
b
an
g
o
’
s
e
s
s
io
n
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
6
,
No
.
1
,
Ma
r
ch
2
0
1
7
:
33
–
4
8
36
T
ab
le
1
.
T
im
e
D
i
v
is
io
n
o
f
C
la
s
s
ical
R
ag
a
s
T
i
me
o
f
P
l
a
y
i
n
g
N
o
t
e
s
T
h
h
a
t
R
a
g
a
U
t
t
o
r
a
n
g
o
(Da
y
se
ssi
o
n
)
4
a
m
-
7
a
m
r
e
,
d
h
a
,
G
a
B
h
a
i
r
a
v
P
u
r
v
i
M
a
r
w
a
R
a
m
u
k
e
l
i
P
a
r
a
sb
a
sa
n
t
L
o
l
i
t
7
a
m
-
1
0
a
m
R
e
,
D
h
a
,
G
a
B
i
l
a
w
a
l
K
a
l
y
a
n
K
h
a
maj
D
e
sh
k
a
r
G
o
u
saran
g
G
a
r
a
1
0
a
m
-
4pm
g
a
,
n
i
K
a
f
i
A
sab
a
r
i
B
h
a
i
r
a
v
i
T
o
d
i
B
h
i
mp
a
l
a
sr
e
e
Jo
u
n
p
u
r
i
B
h
a
i
r
a
v
i
B
i
l
a
s
k
h
a
n
i
-
T
o
d
i
P
u
r
b
a
n
g
o
(
N
i
g
h
t
se
ssi
o
n
)
4pm
-
7
p
m
r
e
,
d
h
a
/
D
h
a
,
G
a
B
h
a
i
r
a
v
P
u
r
v
i
M
a
r
w
a
G
o
u
r
i
P
u
r
v
i
,
S
r
e
e
M
a
r
w
a
7pm
-
1
0
p
m
R
e
,
D
h
a
,
G
a
B
i
l
a
w
a
l
K
a
l
y
a
n
K
h
a
maj
D
u
r
g
a
I
man
k
a
l
y
a
n
K
h
a
maj
,
Jo
y
j
a
y
a
n
t
i
1
0
p
m
-
4
a
m
g
a
,
n
i
K
a
f
i
A
sab
a
r
i
B
h
a
i
r
a
v
i
K
a
f
i
A
n
a
r
a
M
a
l
k
o
s
h
Fig
u
r
e
1
.
T
im
e
C
y
cle
o
f
I
n
d
i
a
n
C
la
s
s
ical
M
u
s
ic
I
n
th
e
n
o
te
-
b
ased
tab
le
m
e
n
t
io
n
ed
as
T
ab
le
2
,
th
er
e
ar
e
all
t
w
e
lv
e
n
o
tes
ar
e
ex
p
lain
e
d
f
o
r
th
e
o
cc
u
r
r
en
ce
s
w
it
h
f
i
v
e
in
d
iv
id
u
al
r
ag
a
ex
p
lai
n
ed
in
d
if
f
er
e
n
t
f
iv
e
s
e
s
s
io
n
s
in
b
o
th
asce
n
d
in
g
a
n
d
d
escen
d
in
g
o
r
d
er
s
.
Fro
m
th
e
s
e
n
o
tes
a
n
al
y
s
i
s
p
atter
n
s
t
h
er
e
ar
e
also
b
e
in
clu
d
ed
th
e
‘
J
ati
’
o
r
s
p
ec
ies
o
f
r
ag
a,
‘
Vad
i
’
o
r
m
o
s
t
s
ig
n
i
f
ica
n
t
n
o
tes
a
n
d
‘
Sa
m
v
ad
i
’
o
r
i
m
m
ed
iate
s
ec
o
n
d
m
o
s
t
s
ig
n
i
f
ica
n
t
n
o
tes
ar
e
d
es
cr
ib
ed
ac
co
r
d
in
g
to
th
e
o
cc
u
r
r
en
ce
s
o
f
th
e
u
n
iq
u
e
n
o
tes o
u
t o
f
t
w
elv
e
i
n
d
i
v
id
u
al
n
o
tes.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
Time
-
B
a
s
ed
R
a
g
a
R
ec
o
mme
n
d
a
tio
n
a
n
d
I
n
fo
r
ma
tio
n
R
etri
ev
a
l o
f..
.
(
S
a
ma
r
jit R
o
y
)
37
T
ab
le
2
.
Sev
er
al
No
te
-
Dis
tr
ib
u
tio
n
E
x
a
m
p
le
s
a
m
o
n
g
R
a
g
as
in
I
n
d
ia
n
C
la
s
s
ical
M
u
s
ic
R
a
g
a
N
o
t
e
s
K
a
l
y
a
n
Jau
n
p
u
r
i
B
a
h
a
r
B
h
a
i
r
a
v
i
B
h
i
mp
a
l
a
sr
e
e
A
r
o
h
a
S
a
/
C
√
√
√
√
r
e
/
C
#
√
R
e
/
D
√
√
g
a
/
D
#
/
E
b
√
√
√
G
a
/
E
√
M
a
/
F
√
√
√
√
ma/
F
#
√
P
a
/
G
√
√
√
√
d
h
a
/
G
#
√
√
D
h
a
/
A
√
√
n
i
/
A
#
/
B
b
√
√
√
√
N
i
/
B
√
√
S
a
/
C
√
√
√
√
√
A
b
o
r
o
h
a
N
i
/
B
√
n
i
/
A
#
/
B
b
√
√
√
√
D
h
a
/
A
√
√
D
h
a
/
G
#
√
√
P
a
/
G
√
√
√
√
√
ma/
F
#
√
M
a
/
F
√
√
√
√
√
G
a
/
E
√
g
a
/
D
#
/
E
b
√
√
√
√
R
e
/
D
√
√
√
√
r
e
/
C
#
√
S
a
/
C
√
√
√
√
√
Jat
i
/
S
p
e
c
i
e
s
6
-
7
6
-
7
6
-
6
7
-
7
6
-
7
V
a
d
i
/
D
o
mi
n
a
n
t
G
a
/
E
S
a
/
C
M
a
/
F
P
a
/
G
M
a
/
F
S
a
mv
a
d
i
/
S
e
mi
-
D
o
mi
n
a
n
t
N
i
/
B
P
a
/
G
S
a
/
C
S
a
/
C
S
a
/
C
T
ab
le
3
.
Fre
q
u
en
c
y
–
R
atio
T
a
b
le
o
f
T
w
el
v
e
No
tes
S
a
/
C
r
e
/
C
#
R
e
/
D
g
a
/
D
#
/
E
b
G
a
/
E
M
a
/
F
ma/
F
#
P
a
/
G
d
h
a
/
G
#
D
h
a
/
A
n
i
/
A
#
/
B
b
N
i
/
B
1
.
0
0
0
1
.
0
5
4
1
.
1
2
5
1
.
1
8
6
1
.
2
5
3
1
.
3
3
3
1
.
4
1
5
1
.
5
0
0
1
.
5
8
0
1
.
6
9
1
1
.
7
7
7
1
.
8
9
3
T
h
e
f
r
eq
u
en
c
y
-
r
at
i
o
is
d
eter
m
i
n
ed
in
C
las
s
ical
M
u
s
ic
a
n
al
y
s
i
s
f
o
r
ev
alu
a
tin
g
t
h
e
p
o
s
s
ib
le
in
i
tial
f
r
eq
u
en
c
y
o
f
n
o
tes
an
d
th
e
f
r
eq
u
en
c
y
r
a
n
g
e
s
in
i
tiated
f
o
r
all
t
w
el
v
e
n
o
tes.
Fre
q
u
e
n
c
y
R
ati
o
s
f
o
r
t
w
el
v
e
n
o
te
s
ex
is
t
i
n
b
et
w
ee
n
t
h
e
in
itia
l
n
o
tes
in
ascen
d
i
n
g
o
r
d
er
a
r
r
an
g
e
m
e
n
t,
‘
Sa
’
o
f
lo
w
er
o
ctav
e
u
p
to
th
e
‘
Sa’
o
f
h
ig
h
er
o
ctav
e
is
2
:1
.
Hen
ce
,
all
th
e
f
r
eq
u
e
n
c
y
-
r
atio
v
al
u
e
s
o
f
ea
ch
an
d
ev
er
y
n
o
te
ar
e
ar
r
an
g
ed
w
i
th
r
esp
ec
t
to
in
itial
‘
Sa’
o
f
s
a
m
e
o
ctav
e
wh
ich
d
ec
eits
b
et
w
ee
n
1
a
n
d
2
in
t
h
e
f
r
eq
u
e
n
c
y
-
r
atio
tab
le.
T
h
e
r
atio
o
f
all
t
h
e
n
o
tes
in
cl
u
d
i
n
g
t
h
e
p
r
eli
m
in
ar
y
‘
Sa
’
its
el
f
u
s
es
to
b
e
s
p
ec
if
i
ed
w
i
th
r
esp
ec
t
to
‘
Sa’
n
o
te.
T
h
e
f
r
eq
u
e
n
c
y
-
r
atio
tab
le,
T
a
b
le
3
a
r
e
illu
s
tr
ated
th
e
ca
lcu
lated
f
r
eq
u
e
n
c
y
r
atio
s
i
n
asce
n
d
in
g
o
r
d
er
o
f
t
w
el
v
e
m
u
s
ical
n
o
tes.
4.
M
E
T
H
O
DS O
F
ANALYS
I
S
W
e
h
av
e
d
i
s
cu
s
s
ed
alr
ea
d
y
th
at
th
e
m
u
s
ic
o
r
au
d
io
s
ig
n
al
ca
n
n
o
t
b
e
ex
p
r
ess
ed
b
y
th
e
lin
ea
r
m
at
h
e
m
a
tical
eq
u
at
io
n
s
o
n
l
y
.
B
ec
au
s
e
t
h
e
li
n
ea
r
s
i
g
n
a
l
ex
p
r
ess
io
n
n
o
r
m
a
ll
y
e
x
p
lai
n
s
b
y
t
h
e
eq
u
atio
n
s
w
h
er
e
th
e
p
ar
a
m
eter
s
o
b
lig
ato
r
y
lik
e,
n
o
v
el
f
r
eq
u
e
n
cie
s
,
th
e
ti
m
e
s
n
ec
es
s
i
tated
f
o
r
co
n
ti
n
u
o
u
s
ti
m
e
s
ig
n
als
o
r
th
e
q
u
an
tit
y
o
f
s
a
m
p
le
s
u
s
ed
f
o
r
d
is
cr
ete
ti
m
e
-
s
i
g
n
a
ls
a
n
d
a
m
p
litu
d
es.
W
h
et
h
er
t
h
e
a
m
p
lit
u
d
es
s
h
o
u
ld
n
o
t
b
e
m
ea
s
u
r
ed
b
y
t
h
e
s
p
ee
ch
r
ec
o
g
n
i
tio
n
s
y
s
te
m
s
h
er
e
to
ex
tr
ac
t
th
e
p
itch
p
er
io
d
s
an
d
th
e
c
o
r
r
esp
o
n
d
in
g
p
itch
v
alu
e
s
o
f
e
v
er
y
n
o
tes
s
in
ce
a
lo
t
o
f
s
in
g
er
s
ca
n
p
la
y
a
p
a
r
ticu
lar
m
u
s
ic
o
r
th
e
r
ag
a
i
n
C
las
s
ical
m
u
s
ic
i
n
v
ar
io
u
s
etiq
u
ette
s
.
So
,
f
o
r
a
p
ar
ticu
lar
m
u
s
ic,
to
tal
a
m
p
lit
u
d
e
in
a
p
la
y
i
n
g
s
es
s
io
n
s
h
o
u
ld
b
e
v
ar
ied
w
it
h
ti
m
e.
Hen
ce
,
h
er
e
in
itiall
y
t
h
e
p
itc
h
a
n
al
y
s
is
o
r
in
it
ial
f
r
eq
u
en
c
y
a
n
al
y
s
is
is
r
eq
u
ir
ed
f
o
r
ex
t
r
ac
tin
g
th
e
m
u
s
ical
f
ea
t
u
r
es
u
n
iq
u
el
y
.
I
n
t
h
is
p
ar
ti
cu
lar
r
ag
a
an
al
y
s
i
s
co
n
ce
p
t
in
th
e
v
er
y
b
e
g
in
n
i
n
g
in
p
u
t
is
cle
ar
l
y
a
s
o
n
g
u
s
u
all
y
5
-
7
A
U
D
A
V
–
S
A
M
P
O
O
R
N
A
6
-
7
S
H
A
D
A
V
–
S
A
M
P
O
O
R
N
A
6
-
6
S
H
A
D
A
V
–
S
A
M
P
O
O
R
N
A
7
-
7
S
A
M
P
O
O
R
N
A
–
S
A
M
P
O
O
R
N
A
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
6
,
No
.
1
,
Ma
r
ch
2
0
1
7
:
33
–
4
8
38
tak
en
i
n
.
w
a
v
f
ile
f
o
r
m
at.
T
h
e
ex
tr
ac
tio
n
s
o
f
p
itc
h
v
al
u
es
f
r
o
m
t
h
e
v
o
ca
l
a
n
d
r
h
y
t
h
m
ic
-
p
a
tter
n
s
o
f
m
u
s
ic
ar
e
v
er
y
i
n
tr
icate
b
ec
au
s
e
t
h
e
in
it
i
al
f
r
eq
u
en
cie
s
n
a
m
el
y
p
itc
h
o
f
m
o
s
t
o
f
th
e
m
u
s
ical
n
o
te
s
an
d
r
h
y
th
m
ic
b
ea
ts
o
f
in
s
tr
u
m
en
t
s
ar
e
s
i
m
ilar
an
d
s
o
litt
le
f
r
ac
tio
n
al
d
is
p
ar
ities
w
h
ich
ar
e
r
o
u
g
h
l
y
i
g
n
o
r
ed
.
Her
e
f
o
r
th
e
p
itch
ex
t
r
ac
tio
n
p
u
r
p
o
s
es,
th
e
W
av
e
Su
r
f
er
So
f
t
w
ar
e
h
a
s
b
ee
n
ap
p
lied
.
T
h
e
f
ile
ty
p
e
s
h
o
u
ld
b
e
m
o
n
o
to
n
ic
f
o
r
ex
tr
ac
tio
n
a
n
d
s
a
m
p
le
e
n
co
d
in
g
h
a
s
b
ee
n
s
et
to
li
n
e
-
1
6
.
T
h
e
s
a
m
p
le
r
ate
i
s
to
b
e
f
ix
ed
i
n
2
2
0
5
0
.
I
t
is
o
p
en
ed
in
a
w
a
v
e
f
o
r
m
s
tr
u
ct
u
r
e.
Fro
m
th
e
S
tr
u
ct
u
r
e
a
p
itch
co
n
to
u
r
o
f
th
a
t so
n
g
i
s
g
en
er
ated
.
I
n
th
is
cir
cu
m
s
tan
ce
th
e
p
itch
f
o
r
m
w
ill
g
i
v
e
al
l
t
h
e
p
it
ch
es
u
s
ed
in
t
h
e
s
o
n
g
,
an
d
th
e
p
itch
d
ata
ar
e
to
b
e
s
a
v
ed
i
n
an
e
x
ce
l
s
h
ee
t.
T
h
e
p
itch
d
ata
ar
e
to
b
e
s
o
r
ted
i
n
a
s
ce
n
d
i
n
g
o
r
d
er
b
u
t
t
h
er
e
o
cc
u
r
s
o
m
e
0
v
alu
e
s
t
h
at
ar
e
to
b
e
d
elete
d
.
T
h
e
r
est
o
f
th
e
p
itch
v
al
u
es
f
r
o
m
th
is
p
itch
co
n
to
u
r
ar
e
u
s
ed
to
b
e
s
et
in
to
th
e
m
id
d
le
o
ctav
e
ac
co
r
d
in
g
to
th
e
f
u
n
d
a
m
en
ta
l
f
r
eq
u
e
n
c
y
r
a
n
g
e
lis
ted
in
th
e
T
ab
le
4
.
I
n
I
n
d
ian
C
la
s
s
ical
M
u
s
ic
t
h
e
t
w
el
v
e
m
u
s
ical
n
o
te
s
ar
e
u
s
ed
to
cr
ea
te
an
d
d
ev
elo
p
a
m
u
s
ic
i
n
clu
d
i
n
g
s
ev
e
n
p
u
r
e
n
o
tes
an
d
f
i
v
e
s
cr
atch
ed
n
o
tes.
B
y
t
h
e
f
u
n
d
a
m
e
n
tal
f
r
eq
u
en
c
y
r
an
g
e
an
a
l
y
s
is
T
ab
le
4
th
e
p
itch
v
al
u
e
s
et
h
a
s
b
ee
n
d
is
tr
ib
u
ted
in
to
t
w
el
v
e
g
r
o
u
p
s
ac
co
r
d
in
g
to
th
e
s
eq
u
en
ce
s
o
f
th
e
ele
m
en
tar
y
n
o
te
s
tr
u
ctu
r
e
s
.
Fro
m
th
e
s
e
t
w
el
v
e
g
r
o
u
p
s
o
f
b
asic
n
o
te
-
s
t
r
u
ctu
r
es,
th
e
h
i
g
h
est
o
cc
u
r
r
en
ce
o
f
p
r
im
ar
y
f
r
eq
u
en
c
y
r
an
g
e
h
a
s
b
ee
n
in
itia
te
d
.
T
h
is
h
ig
h
es
t
f
r
eq
u
e
n
c
y
v
alu
e
o
u
t
o
f
t
w
el
v
e
f
r
eq
u
en
c
y
r
an
g
es
i
s
k
n
o
w
n
as
th
e
m
o
s
t
s
i
g
n
i
f
ican
t
n
o
te
o
r
‘
Vad
i’
in
I
n
d
ia
n
C
lass
ic
al
m
u
s
ic.
Af
ter
th
e
ev
alu
a
tio
n
o
f
th
e
m
o
s
t si
g
n
if
ic
an
t
n
o
te,
th
e
r
est
s
o
f
th
e
f
r
eq
u
en
c
y
r
a
n
g
e
s
ar
e
ev
al
u
ated
b
y
t
h
e
f
r
eq
u
en
c
y
-
r
a
tio
s
o
f
f
u
n
d
a
m
e
n
tal
t
w
elv
e
n
o
tes
in
I
C
M.
T
h
e
m
u
s
ical
p
atter
n
s
ar
e
esti
m
ated
h
er
e
f
o
r
r
ec
o
g
n
izi
n
g
th
e
n
o
tes
b
y
u
s
i
n
g
t
h
eir
co
r
r
esp
o
n
d
in
g
p
itc
h
v
al
u
e
s
.
He
n
ce
t
h
e
m
atc
h
i
n
g
b
et
w
ee
n
e
v
al
u
ated
r
est
s
o
f
t
h
e
f
r
eq
u
e
n
c
y
r
an
g
es
w
it
h
t
h
e
f
u
n
d
a
m
e
n
tal
f
r
eq
u
e
n
c
y
r
an
g
es
o
f
ac
tu
al
m
u
s
ic.
T
h
is
h
a
s
to
b
e
s
ee
n
w
h
et
h
er
th
e
m
atc
h
ed
n
o
tes
ar
e
esti
m
ated
as
u
s
ed
n
o
tes o
f
t
h
e
in
p
u
t
au
d
io
f
ile.
I
f
t
h
e
m
atc
h
in
g
a
n
al
y
s
i
s
h
as
s
u
c
ce
s
s
f
u
ll
y
d
o
n
e,
t
h
en
t
h
is
h
a
s
to
b
e
s
ee
n
t
h
at
w
h
et
h
er
t
h
is
u
s
ed
n
o
tes
ar
e
g
u
es
s
ti
m
ated
p
r
ec
is
el
y
w
it
h
th
e
ti
m
e
-
b
ase
d
n
o
te
an
d
t
h
eir
co
r
r
esp
o
n
d
in
g
f
ea
tu
r
e
s
tab
le
an
d
r
ag
a
ti
m
e
-
b
ased
cir
cle.
T
h
e
en
tire
f
lo
w
o
f
t
h
e
ac
ti
v
ities
to
e
x
tr
ac
t
t
h
e
f
r
eq
u
en
c
ies,
ca
lc
u
lati
n
g
m
id
d
le
o
ctav
e,
s
ea
r
c
h
in
g
i
n
lo
g
ica
l
w
a
y
f
o
r
Mo
s
t
S
ig
n
i
f
ica
n
t
N
o
te
o
f
a
p
ar
ticu
lar
v
o
ca
l
p
er
f
o
r
m
a
n
ce
,
m
atc
h
in
g
n
o
tes
e
s
ti
m
atio
n
s
t
h
at
e
x
i
s
t
i
n
a
r
ag
a
p
er
f
o
r
m
an
ce
etc.
ar
e
b
r
ief
l
y
ex
p
lai
n
ed
i
n
Fig
u
r
e
2
as
a
f
lo
w
-
c
h
ar
t.
T
h
e
an
al
y
s
is
u
n
d
er
th
is
m
etic
u
lo
u
s
co
n
ce
p
t
h
as
b
ee
n
ev
id
en
tl
y
v
i
s
u
alize
d
t
h
at
in
t
h
i
s
ap
p
r
o
ac
h
th
e
m
u
s
ical
p
atter
n
s
ar
e
o
p
ti
m
all
y
d
i
s
ti
n
g
u
is
h
ed
a
m
o
n
g
t
h
er
e
n
o
tes
d
esc
r
ib
ed
in
t
h
e
r
ag
a
class
i
f
icatio
n
an
d
r
ec
o
g
n
itio
n
.
T
h
is
m
et
h
o
d
is
also
ap
p
licab
le
f
o
r
th
e
m
u
s
ical
n
o
tes a
n
d
p
atter
n
class
i
f
icat
io
n
s
.
T
ab
le
4
.
Mu
s
ical
No
tes an
d
th
eir
Fu
n
d
a
m
e
n
tal
Fre
q
u
e
n
c
y
R
an
g
e
s
W
e
st
e
r
n
N
o
t
e
s
I
n
d
i
a
n
N
o
t
e
s
F
r
e
q
u
e
n
c
y
R
a
n
g
e
A
D
h
a
2
1
5
.
5
–
2
2
6
.
5
A#
ni
2
2
6
.
6
–
2
3
9
.
9
B
Ni
2
4
0
.
0
–
2
5
2
.
5
C
Sa
2
5
2
.
6
–
2
6
9
.
5
C#
re
2
6
9
.
6
–
2
8
5
.
3
D
Re
2
8
5
.
4
–
3
0
2
.
3
D#
ga
3
0
2
.
4
–
3
2
0
.
3
E
Ga
3
2
0
.
4
–
3
3
9
.
3
F
Ma
3
3
9
.
4
–
3
5
9
.
4
F#
ma
3
5
9
.
5
–
3
8
0
.
9
G
Pa
3
8
1
.
0
–
4
0
3
.
5
G#
d
h
a
4
0
3
.
6
–
4
2
7
.
5
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
Time
-
B
a
s
ed
R
a
g
a
R
ec
o
mme
n
d
a
tio
n
a
n
d
I
n
fo
r
ma
tio
n
R
etri
ev
a
l o
f..
.
(
S
a
ma
r
jit R
o
y
)
39
Fig
u
r
e
2
.
Flo
w
-
C
h
ar
t
f
o
r
Dete
r
m
in
a
tio
n
o
f
R
ag
a
f
r
o
m
a
C
o
m
p
lete
Mu
s
ic
4
.
1
.
F
e
a
t
ures E
x
t
ra
ct
io
n
Fo
r
ti
m
e
-
b
a
s
ed
class
i
f
icatio
n
o
f
r
ag
a,
th
r
ee
b
asic
f
ea
tu
r
e
s
ar
e
ex
tr
ac
ted
in
t
h
e
ex
p
er
i
m
e
n
t
–
1.
Nu
m
b
er
o
f
n
o
te
s
(
s
w
ar
as)
p
r
esen
t
i
n
a
r
ag
a
f
r
o
m
th
e
1
2
n
o
tes
o
f
I
n
d
ia
n
cl
ass
ical
M
u
s
ic
(
Usi
n
g
w
a
v
es
u
r
f
er
….
)
.
2.
T
h
e
co
m
b
in
a
tio
n
o
f
n
o
tes o
f
a
p
ar
ticu
lar
r
ag
a.
3.
T
h
e
Mo
s
t Sig
n
if
ican
t N
o
te
(
V
ad
i S
w
ar
)
w
h
ich
i
s
u
s
ed
m
o
s
t
f
r
eq
u
en
tl
y
i
n
a
r
ag
a
(
Us
in
g
w
a
v
es
u
r
f
er
….
.
)
.
C
o
m
b
i
n
atio
n
o
f
n
o
te
s
o
r
s
w
a
r
a
co
m
b
in
at
io
n
i
s
th
e
b
asic
s
et
o
f
n
o
tes
w
h
ic
h
i
s
u
s
ed
in
a
p
ar
ticu
lar
r
ag
a.
T
h
e
co
m
b
i
n
atio
n
o
f
n
o
te
s
ca
n
b
e
r
ep
r
esen
ted
b
y
b
in
ar
y
b
its
.
T
h
e
b
in
ar
y
s
eq
u
e
n
ce
s
o
f
b
its
ar
e
co
n
v
er
ted
in
to
d
ec
i
m
al.
Fo
r
ex
a
m
p
le,
r
ag
a
Kal
y
an
u
s
e
s
th
e
n
o
te
’
s
co
m
b
i
n
atio
n
–
s
a
R
e
Ga
m
a
p
a
Dh
a
Ni.
So
,
th
e
b
in
ar
y
b
it
s
eq
u
en
ce
w
h
ic
h
r
ep
r
esen
t
s
th
e
co
m
b
i
n
atio
n
o
f
n
o
tes i
s
i
n
th
e
tab
le.
T
ab
le
5
.
C
o
m
b
in
a
tio
n
o
f
N
o
te
s
in
t
h
e
R
a
g
a
‘
Kal
y
a
n
’
/ T
h
h
at
Kal
y
a
n
S
a
/
C
r
e
/
C#
R
e
/
D
g
a
/
D
#
/
E
b
G
a
/
E
M
a
/
F
ma/
F
#
P
a
/
G
d
h
a
/
G
#
D
h
a
/
A
n
i
/
A
#
/
B
b
N
i
/
B
1
0
1
0
1
0
1
1
0
1
0
1
T
h
e
b
in
ar
y
s
eq
u
en
ce
is
–
1
0
1
0
1
1
0
1
0
1
0
1
.
T
h
e
n
o
te
‘
Sa’
is
p
r
esen
t
in
e
v
er
y
r
ag
a,
s
o
‘
Sa’
c
an
b
e
o
m
it
ted
f
r
o
m
th
e
b
in
ar
y
b
it
s
eq
u
e
n
ce
.
B
y
o
m
itti
n
g
‘
Sa
’
,
th
e
r
es
u
lta
n
t
b
in
ar
y
b
it
s
eq
u
e
n
ce
is
–
0
1
0
1
1
0
1
0
1
0
1
w
h
o
s
e
d
ec
i
m
a
l
v
alu
e
i
s
6
9
3
.
B
y
th
i
s
w
a
y
d
ec
i
m
al
v
a
lu
e
s
o
f
n
o
te
’
s
co
m
b
in
at
io
n
f
o
r
all
r
ag
as a
r
e
ca
lc
u
lated
.
4
.
2
.
E
x
peri
m
ent
Det
a
ils
4
.
2
.
1
.
T
he
neura
l net
w
o
rk
T
h
e
class
if
icatio
n
o
f
r
a
g
as
b
a
s
ed
o
n
th
eir
p
la
y
i
n
g
ti
m
e
is
d
o
n
e
b
y
u
s
in
g
a
t
w
o
la
ye
r
feed
-
fo
r
w
a
r
d
n
eu
r
al
n
et
w
o
r
k
w
h
ic
h
is
s
h
o
wn
b
elo
w
:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
6
,
No
.
1
,
Ma
r
ch
2
0
1
7
:
33
–
4
8
40
Fig
u
r
e
3
.
T
h
e
C
o
m
p
lete
Net
o
f
th
e
P
r
o
p
o
s
ed
A
n
al
y
s
is
T
h
e
Fig
u
r
e
3
is
t
h
e
co
m
p
lete
n
et
o
f
n
e
u
r
al
n
et
w
o
r
k
w
h
ic
h
h
as
o
n
e
in
p
u
t
la
ye
r
,
o
n
e
o
u
tp
u
t
la
ye
r
an
d
o
n
e
la
y
er
b
e
t
w
ee
n
t
h
e
in
p
u
t
la
y
er
an
d
t
h
e
o
u
tp
u
t
la
y
er
w
h
ic
h
is
ca
l
led
th
e
h
id
d
en
la
ye
r
.
T
h
e
i
n
p
u
t
la
y
er
d
o
es
n
o
t
ta
k
e
p
ar
t
i
n
an
y
co
m
p
u
t
in
g
r
eg
ar
d
i
n
g
clas
s
if
icatio
n
.
T
h
e
co
m
p
u
tat
io
n
o
cc
u
r
s
o
n
l
y
b
y
th
e
n
e
u
r
o
n
s
p
r
ese
n
t
in
t
h
e
h
id
d
en
la
y
er
an
d
t
h
e
o
u
tp
u
t
la
y
er
.
So
t
h
e
n
e
u
r
al
n
et
wo
r
k
is
tw
o
la
ye
r
s
Neu
r
al
Net
wo
r
k
.
T
h
e
in
p
u
ts
ar
e
f
ed
f
r
o
m
in
p
u
t
la
y
er
to
th
e
h
i
d
d
en
lay
er
an
d
th
e
n
to
th
e
o
u
tp
u
t
la
y
er
.
So
th
e
n
et
w
o
r
k
is
f
ee
d
fo
r
w
a
r
d
n
eu
r
al
n
et
w
o
r
k
.
T
h
e
Fig
u
r
e
also
s
h
o
w
s
th
at
t
h
e
n
et
w
o
r
k
u
s
es
s
ig
m
o
id
fu
n
ctio
n
as
th
e
ac
tiv
atio
n
f
u
n
ctio
n
in
b
o
th
o
f
th
e
h
id
d
en
a
n
d
th
e
o
u
tp
u
t
la
y
er
.
T
h
r
ee
f
ea
tu
r
es
o
f
ea
ch
s
a
m
p
le
ar
e
ap
p
lied
o
n
th
e
in
p
u
t
la
y
er
o
f
th
e
n
e
u
r
al
n
et
w
o
r
k
a
n
d
h
en
ce
t
h
e
i
n
p
u
t
l
a
y
er
h
a
s
t
h
r
ee
n
eu
r
o
n
s
.
T
h
e
o
u
tp
u
t
la
y
er
h
a
s
t
w
o
n
e
u
r
o
n
s
as
th
e
in
p
u
t
d
ata
h
a
v
e
to
b
e
class
if
ied
i
n
to
t
w
o
cla
s
s
e
s
–
P
u
r
v
an
g
o
an
d
Uttar
an
g
o
.
T
h
e
n
u
m
b
er
o
f
h
id
d
en
la
y
er
m
a
y
v
ar
y
ac
co
r
d
in
g
to
t
h
e
ap
p
licatio
n
r
eq
u
ir
e
m
e
n
t
a
n
d
ac
cu
r
ac
y
r
eq
u
ir
e
m
en
t.
I
n
m
o
s
t
ap
p
licatio
n
,
o
n
e
h
id
d
en
la
y
er
i
s
s
u
f
f
ic
ien
t.
I
n
t
h
is
e
x
p
er
i
m
e
n
t,
t
h
e
h
id
d
en
la
y
er
h
as
1
5
n
eu
r
o
n
s
o
r
co
m
p
u
tatio
n
n
o
d
es
w
h
ic
h
also
m
a
y
v
ar
y
ac
co
r
d
in
g
to
ac
cu
r
ac
y
r
eq
u
ir
e
m
e
n
t.
4
.
2
.
2
.
T
ra
ini
ng
a
nd
T
esti
ng
T
o
tal
9
2
s
am
p
les
ar
e
co
llected
f
r
o
m
2
3
r
ag
as
f
o
r
th
i
s
class
if
i
ca
tio
n
ex
p
er
i
m
e
n
t.
T
h
e
d
ataset
co
n
tain
s
th
e
n
u
m
er
ical
v
al
u
es
o
f
t
h
r
ee
f
ea
t
u
r
es
o
f
ea
ch
s
a
m
p
le.
F
o
r
t
r
ain
in
g
,
7
0
%
o
f
to
tal
s
a
m
p
les
h
av
e
u
s
ed
an
d
1
5
%
s
a
m
p
les
h
a
v
e
u
s
ed
f
o
r
test
in
g
.
R
est
o
f
1
5
%
s
a
m
p
les
h
a
v
e
u
s
ed
f
o
r
v
alid
atio
n
p
u
r
p
o
s
e.
T
h
e
w
ei
g
h
ts
ar
e
g
en
er
ated
b
y
t
h
e
s
y
s
te
m
ac
co
r
d
in
g
to
th
e
g
i
v
en
i
n
p
u
t.
T
h
e
f
o
llo
w
i
n
g
tab
le
6
s
h
o
w
s
t
h
e
f
ea
t
u
r
es o
f
1
5
r
a
g
as:
T
ab
le
6
.
Data
s
et
o
f
T
h
r
ee
B
asi
c
Featu
r
es o
f
t
h
e
R
a
g
a
N
o
.
R
a
g
a
F
e
a
t
u
r
e
s C
a
l
c
u
l
a
t
e
d
N
o
.
o
f
N
o
t
e
s
C
o
mb
i
n
a
t
i
o
n
o
f
n
o
t
e
s
M
o
st
S
i
g
n
i
f
i
c
a
n
t
N
o
t
e
(
V
a
d
i
)
1
Y
a
man
7
6
9
3
3
2
0
2
V
i
l
a
v
a
l
7
7
2
7
2
1
6
3
K
a
f
i
7
8
5
4
3
8
1
4
B
i
h
a
g
7
7
2
5
3
2
0
5
A
sab
a
r
i
7
8
5
8
4
0
4
6
B
h
a
i
r
a
v
i
7
1
3
7
0
3
6
0
7
D
e
sh
7
7
2
7
2
8
5
8
B
a
g
e
sh
r
i
7
8
5
4
3
6
0
9
D
u
r
g
a
5
5
9
6
3
6
0
10
Jo
u
n
p
u
r
i
7
8
5
8
4
0
4
11
M
a
l
k
o
s
h
5
3
3
0
3
6
0
12
K
e
d
a
r
7
7
5
7
3
6
0
13
P
a
t
d
w
i
p
7
8
5
3
3
8
1
14
B
h
a
i
r
a
v
7
1
2
4
1
4
0
4
15
B
h
u
p
a
l
i
5
6
6
0
3
2
0
5.
P
E
RF
O
RM
ANCE AN
AL
YS
I
S
5
.
1
.
Co
nfusi
o
n
M
a
t
ri
x
P
er
f
o
r
m
a
n
ce
o
f
a
c
lass
if
icatio
n
ca
n
b
e
b
est
s
ee
n
w
it
h
t
h
e
h
elp
o
f
th
e
C
o
n
f
u
s
io
n
Ma
tr
i
x
.
C
o
n
f
u
s
io
n
Ma
tr
ix
co
n
tai
n
s
th
e
i
n
f
o
r
m
ati
o
n
ab
o
u
t
ac
cu
r
ac
y
o
f
th
e
clas
s
if
icatio
n
b
et
w
ee
n
th
e
ac
t
u
al
an
d
p
r
ed
icted
class
es
d
o
n
e
b
y
th
e
clas
s
if
icatio
n
s
y
s
te
m
.
T
h
e
C
o
n
f
u
s
i
o
n
m
a
t
r
ix
h
e
lp
s
to
ev
a
lu
ate
th
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
class
i
f
icatio
n
s
y
s
te
m
u
s
i
n
g
t
h
e
d
ata
p
r
esen
t
i
n
t
h
e
m
atr
i
x
.
A
cc
o
r
d
in
g
to
t
h
e
C
o
n
f
u
s
i
o
n
Ma
tr
ix
o
f
th
e
class
i
f
icatio
n
d
o
n
e
b
y
t
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
,
ea
c
h
co
lu
m
n
r
ep
r
esen
ts
t
h
e
T
ar
g
et
o
r
A
ct
u
al
class
a
n
d
ea
ch
r
o
w
r
ep
r
esen
ts
t
h
e
O
u
tp
u
t
clas
s
o
r
P
r
ed
icted
class
.
T
h
er
e
ar
e
t
w
o
class
es
f
o
r
clas
s
i
f
icatio
n
-
a)
P
u
r
va
n
g
a
(
cla
s
s
1
)
b
)
Utta
r
a
n
g
a
(
cla
s
s
2
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
Time
-
B
a
s
ed
R
a
g
a
R
ec
o
mme
n
d
a
tio
n
a
n
d
I
n
fo
r
ma
tio
n
R
etri
ev
a
l o
f..
.
(
S
a
ma
r
jit R
o
y
)
41
Fig
u
r
e
4
.
C
o
n
f
u
s
io
n
Ma
tr
ices
Fig
u
r
e
1
s
h
o
w
s
t
h
e
f
o
u
r
co
n
f
u
s
io
n
m
a
tr
ices
–
T
r
ain
in
g
,
Vali
d
atio
n
,
T
est
an
d
A
ll
co
n
f
u
s
io
n
m
atr
ices.
E
ac
h
r
o
w
o
f
th
e
m
a
tr
ix
r
ep
r
e
s
en
t
s
t
h
e
O
u
tp
u
t
o
r
p
r
ed
icted
class
an
d
ea
c
h
co
l
u
m
n
r
ep
r
e
s
en
t
s
t
h
e
T
ar
g
et
o
r
A
ct
u
al
clas
s
.
T
h
e
g
r
ee
n
s
q
u
ar
es
d
en
o
te
th
e
n
u
m
b
er
o
f
s
a
m
p
les
w
h
ic
h
ar
e
co
r
r
ec
tly
cla
s
s
i
f
i
ed
b
y
th
e
cla
s
s
i
f
ier
,
an
d
th
e
r
ed
s
q
u
ar
es
d
e
n
o
te
t
h
e
n
u
m
b
er
o
f
s
a
m
p
les
w
h
ic
h
a
r
e
in
co
r
r
ec
tl
y
clas
s
i
f
ied
.
T
h
e
b
lu
e
s
q
u
ar
es
d
en
o
te
th
e
o
v
er
all
p
er
ce
n
ta
g
e
o
f
co
r
r
ec
t
class
if
ica
tio
n
.
I
t
i
s
a
s
s
u
m
e
d
th
at
i
f
t
h
e
o
u
tco
m
e
o
f
t
h
e
c
lass
i
f
icatio
n
i
s
t
h
e
m
e
m
b
er
o
f
cla
s
s
Nig
h
t
Ses
s
io
n
(
clas
s
1
)
,
th
e
n
th
e
y
w
i
ll
b
e
t
r
ea
ted
as
n
e
g
ati
v
e
o
u
tco
m
e
a
n
d
if
t
h
e
o
u
tco
m
e
o
f
th
e
clas
s
i
f
icatio
n
i
s
th
e
m
e
m
b
er
o
f
class
Uttar
a
n
g
a
(
clas
s
2
)
,
th
e
y
w
i
ll b
e
tr
ea
ted
as p
o
s
itiv
e
o
u
tco
m
e.
5
.
2
.
T
ra
ini
ng
Co
nfusi
o
n M
a
t
rix
T
r
ain
in
g
C
o
n
f
u
s
io
n
m
atr
i
x
i
n
Fig
u
r
e
5
d
e
m
o
n
s
tr
ates
th
e
p
e
r
ce
n
tag
e
o
f
co
r
r
ec
t
class
i
f
ica
ti
o
n
d
u
r
in
g
T
r
ain
in
g
b
y
t
h
e
s
a
m
p
les.
I
n
t
h
is
e
x
p
er
i
m
e
n
t,
to
tal
9
2
s
a
m
p
les
o
f
2
3
r
ag
as
ar
e
co
llected
an
d
7
0
%
o
f
th
e
to
tal
s
a
m
p
les ar
e
u
s
ed
f
o
r
tr
ain
i
n
g
i
.
e.
(
7
0
% o
f
9
2
)
o
r
6
4
s
a
m
p
les
ar
e
u
s
ed
f
o
r
tr
ain
i
n
g
.
Fig
u
r
e
5
.
T
r
ain
in
g
C
o
n
f
u
s
i
o
n
Ma
tr
ix
N
o
.
o
f
C
o
r
r
e
c
t
l
y
C
l
a
s
s
i
f
i
e
d
S
a
m
p
l
e
s
N
o
.
o
f
I
n
c
o
r
r
e
c
t
l
y
C
l
a
s
s
i
f
i
e
d
S
a
m
p
l
e
s
C
o
r
r
e
c
t
c
l
a
s
s
i
f
i
c
a
t
i
o
n
p
e
r
c
e
n
t
a
g
e
P
r
e
d
i
c
t
e
d
o
u
t
c
o
m
e
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
6
,
No
.
1
,
Ma
r
ch
2
0
1
7
:
33
–
4
8
42
T
ab
le
7
.
R
esu
lt Set
o
f
T
r
ain
i
n
g
C
o
n
f
u
s
io
n
Ma
tr
ix
T
r
a
i
n
i
n
g
C
o
n
f
u
s
i
o
n
M
a
t
r
i
x
N
o
.
o
f
S
a
mp
l
e
s
N
a
me
o
f
C
l
a
ss
N
o
.
o
f
samp
l
e
s
a
ssi
g
n
e
d
t
o
C
l
a
ss
TP
TN
FP
FN
P
e
r
c
e
n
t
a
g
e
o
f
S
a
mp
l
e
s
p
r
e
d
i
c
t
e
d
P
e
r
c
e
n
t
a
g
e
o
f
A
ssi
g
n
e
d
S
a
mp
l
e
s
O
v
e
r
a
l
l
P
e
r
c
e
n
t
a
g
e
64
1
37
0
37
0
0
1
0
0
8
8
.
1
9
2
.
2
2
22
22
0
0
5
8
1
.
5
1
0
0
T
P
:
T
r
u
e
P
o
si
t
i
v
e
;
T
N
:
Tr
u
e
N
e
g
a
t
i
v
e
;
F
P
:
F
a
l
se
P
o
si
t
i
v
e
;
F
N
:
F
a
l
se
N
e
g
a
t
i
v
e
A
cc
o
r
d
in
g
to
th
e
T
r
ain
in
g
C
o
n
f
u
s
io
n
Ma
tr
ix
an
d
o
u
tco
m
e
s
r
ep
r
esen
ted
co
r
r
esp
o
n
d
in
g
l
y
i
n
F
ig
u
r
e
5
an
d
T
ab
le
7
,
3
7
s
am
p
le
s
ar
e
co
r
r
ec
tly
a
s
s
i
g
n
ed
to
class
1
.
T
h
er
ef
o
r
e,
T
r
u
e
Neg
ati
v
e
Val
u
e
is
3
7
.
0
s
a
m
p
le
s
ar
e
ass
ig
n
ed
to
clas
s
2
.
T
h
er
ef
o
r
e,
Fals
e
P
o
s
iti
v
e
Val
u
e
is
0
.
1
0
0
*
(
3
7
/3
7
)
=
1
0
0
%
o
f
c
lass
1
s
a
m
p
les
ar
e
p
r
ed
icted
as
class
1
co
r
r
ec
tly
.
2
2
s
am
p
le
s
ar
e
co
r
r
ec
tly
a
s
s
i
g
n
ed
to
class
2
.
T
h
er
ef
o
r
e,
T
r
u
e
P
o
s
itiv
e
Valu
e
i
s
2
2
.
5
s
am
p
les
ar
e
as
s
ig
n
ed
i
n
co
r
r
ec
tly
to
clas
s
1
b
u
t
t
h
e
y
a
r
e
m
e
m
b
er
o
f
cla
s
s
2
.
T
h
er
ef
o
r
e,
Fals
e
Neg
ati
v
e
Valu
e
is
5
.
1
0
0
*
(
2
2
/2
7
)
=
8
1
.
5
%
o
f
cla
s
s
2
s
a
m
p
les
ar
e
p
r
ed
icted
as
clas
s
2
co
r
r
ec
tl
y
.
1
0
0
*
(
5
/2
7
)
=
1
8
.
5
%
o
f
c
lass
2
s
a
m
p
les ar
e
p
r
ed
icted
in
co
r
r
ec
tl
y
as c
la
s
s
1
.
1
0
0
*
(
3
7
/
4
2
)
=
8
8
.
1
% o
f
ass
ig
n
m
e
n
t o
f
class
1
ar
e
co
r
r
ec
t.
1
0
0
*
(
2
2
/2
2
)
=
1
0
0
%
o
f
ass
i
g
n
m
e
n
t
o
f
cla
s
s
2
ar
e
co
r
r
ec
t.
T
h
e
p
er
ce
n
tag
e
o
f
o
v
er
all
co
r
r
ec
tl
y
class
if
ied
d
ata
f
o
r
tr
ain
i
n
g
i
s
(
(
3
7
+2
2
)
/ (
3
7
+
5
+2
2
)
)
*
1
0
0
=
9
2
.
2
%.
5
.
3
.
Va
lid
a
t
i
o
n Co
nfusi
o
n M
a
t
rix
Valid
atio
n
C
o
n
f
u
s
io
n
m
atr
i
x
in
Fi
g
u
r
e
6
d
is
p
la
y
s
t
h
e
p
er
ce
n
tag
e
o
f
co
r
r
ec
t
class
if
icatio
n
d
u
r
in
g
Valid
atio
n
o
f
t
h
e
s
a
m
p
les.
I
n
t
h
is
ex
p
er
i
m
en
t,
1
5
% o
f
to
tal
s
a
m
p
les
ar
e
u
s
ed
f
o
r
v
al
id
atio
n
i.e
.
(
1
5
% o
f
9
2
)
o
r
1
4
s
am
p
le
s
ar
e
u
s
ed
f
o
r
v
al
id
atio
n
.
Fig
u
r
e
6
.
Valid
atio
n
C
o
n
f
u
s
io
n
Ma
tr
ix
T
ab
le
8
.
R
esu
lt Set
o
f
T
r
ain
i
n
g
C
o
n
f
u
s
io
n
Ma
tr
ix
V
a
l
i
d
a
t
i
o
n
C
o
n
f
u
s
i
o
n
M
a
t
r
i
x
N
o
.
o
f
S
a
mp
l
e
s
N
a
me
o
f
C
l
a
ss
N
o
.
o
f
samp
l
e
s
a
ssi
g
n
e
d
t
o
C
l
a
ss
TP
TN
FP
FN
P
e
r
c
e
n
t
a
g
e
o
f
S
a
mp
l
e
s
p
r
e
d
i
c
t
e
d
P
e
r
c
e
n
t
a
g
e
o
f
A
ssi
g
n
e
d
S
a
mp
l
e
s
O
v
e
r
a
l
l
P
e
r
c
e
n
t
a
g
e
14
1
10
0
10
0
0
1
0
0
9
0
.
9
9
2
.
9
2
3
3
0
0
1
8
1
.
5
1
0
0
T
P
:
T
r
u
e
P
o
si
t
i
v
e
;
T
N
:
Tr
u
e
N
e
g
a
t
i
v
e
;
F
P
:
F
a
l
se
P
o
si
t
i
v
e
;
F
N
:
F
a
l
se
N
e
g
a
t
i
v
e
A
cc
o
r
d
in
g
to
th
e
Valid
atio
n
C
o
n
f
u
s
io
n
Ma
tr
i
x
an
d
o
u
tco
m
es
ch
ar
ac
ter
ized
in
Fig
u
r
e
6
a
n
d
T
a
b
le
8
co
r
r
esp
o
n
d
in
g
l
y
,
1
0
s
a
m
p
le
s
ar
e
co
r
r
ec
tly
as
s
i
g
n
ed
to
class
1
.
T
h
er
ef
o
r
e,
T
r
u
e
Neg
at
iv
e
Val
u
e
is
1
0
.
0
s
a
m
p
les
ar
e
as
s
ig
n
ed
to
class
2
.
T
h
er
ef
o
r
e,
Fals
e
P
o
s
itiv
e
Valu
e
is
0
.
1
0
0
*
(
1
0
/1
0
)
=
1
0
0
%
o
f
class
1
s
a
m
p
les
ar
e
p
r
ed
icted
as
clas
s
1
p
r
o
p
er
l
y
.
3
s
a
m
p
les
ar
e
co
r
r
ec
tl
y
as
s
ig
n
ed
to
cla
s
s
2
.
T
h
er
ef
o
r
e,
T
r
u
e
P
o
s
itiv
e
Val
u
e
is
3
.
1
s
a
m
p
le
is
as
s
i
g
n
ed
i
n
c
o
r
r
ec
tly
to
clas
s
1
b
u
t
t
h
e
y
ar
e
m
e
m
b
er
o
f
cla
s
s
2
.
T
h
er
ef
o
r
e,
Fals
e
Neg
ati
v
e
Valu
e
i
s
1
.
1
0
0
*
(
3
/4
)
=
8
1
.
5
%
o
f
class
2
s
a
m
p
le
s
ar
e
p
r
ed
icted
as c
lass
2
s
u
itab
l
y
.
1
0
0
*
(
1
/
4
)
=
1
8
.
5
% o
f
class
2
s
a
m
p
les
ar
e
p
r
ed
icted
in
co
r
r
ec
tl
y
as
clas
s
1
.
1
0
0
*
(
1
0
/1
1
)
=
9
0
.
9
%
o
f
ass
ig
n
m
en
t
o
f
class
1
ar
e
co
r
r
ec
t.
Evaluation Warning : The document was created with Spire.PDF for Python.